Would you ask AI who to vote for? | Elections and Technology Reader Issue No. 10 | Asian Network for Free Elections

August 18, 2026 - 16:36

The latest edition of the Elections and Technology Reader takes a close look at a growing trend: voters turning to artificial intelligence chatbots for help deciding who to support at the ballot box. The publication, produced by the Asian Network for Free Elections, explores how these tools are being used as both information sources and informal candidate evaluators, and what that means for the integrity of elections across the region.

Chatbots like ChatGPT and similar large language models are now capable of summarizing party platforms, comparing policy positions, and even answering direct questions about candidates. For many voters, especially younger ones, this feels like a natural extension of how they already use technology. Instead of wading through lengthy manifestos or tracking down multiple news articles, they can simply ask an AI to break down the options.

But the report raises serious concerns. AI systems are not neutral. They are trained on vast amounts of data that may contain biases, and they can produce confident but incorrect answers. When it comes to something as important as an election, a small error or a subtle slant in a response could mislead a voter. The reader highlights cases where chatbots have given outdated or incomplete information about voting procedures, and other instances where they have struggled to accurately represent the positions of lesser-known candidates.

There is also the question of accountability. If a chatbot gives bad advice, who is responsible? The company that built the model? The platform that hosts it? Or the voter who chose to rely on it? Current regulations in most countries do not clearly address this. ANFREL suggests that election management bodies need to start thinking about how to guide the use of AI in the electoral process, not by banning it outright, but by setting standards for transparency and accuracy.

The issue also looks at the positive side. In places where traditional media is limited or where political information is hard to access, chatbots could help level the playing field. They can provide quick answers in multiple languages and help voters with disabilities or low literacy. The key, according to the report, is to make sure that these tools are built and used in a way that supports informed choice rather than undermining it.

As elections become more digital, the question is no longer whether voters will use AI, but how well the systems will serve them. This reader serves as a timely reminder that technology is not a replacement for civic education and critical thinking. It is a tool, and like any tool, it can be used well or misused. The hope is that by highlighting these issues now, election watchdogs and tech developers can work together to keep the process fair and trustworthy.